Table of Contents, Summaries of Nursing

Download Table of Contents and more Summaries Nursing in PDF only on Docsity! Stanford Quality Improvement/Patient Safety Symposium 2015 Monday, June 8, 2015 at 3:30 PM – 6:30 PM Li Ka Shing Center (LKSC), Berg Hall, LK 230A Revised 06/04/2015 Page 1 of 21 Table of Contents Agenda .................................................................................. 2 Keynote Speaker ................................................................. 3 Poster Map ........................................................................... 4 Abstracts .............................................................................. 5 Stanford Quality Improvement/Patient Safety Symposium 2015 Monday, June 8, 2015 at 3:30 PM – 6:30 PM Li Ka Shing Center (LKSC), Berg Hall, LK 230A Revised 06/04/2015 Page 2 of 21 Agenda Time Topic 12:00 PM – 3:00 PM Poster Setup  Pushpins will be provided. 3:30 PM – 5:00 PM Poster Session and Judging 5:00 PM – 5:15 PM Introduction/Keynote Session Terry Platchek, MD Medical Director, Performance Improvement Stanford Children’s Health 5:15 PM – 5:30 PM Resident Safety Council : “Improving consultations: a pilot study of a brief consult note to standardize emergency room and consultant communication” 5:30 PM – 5:45 PM Resident Safety Council : “Safety Culture at Stanford: Improving physician utilization of an adverse event reporting system” 5:45 PM – 6:00 PM Resident Safety Council : “Barriers to Effective RN-MD Communication at Stanford” 6:00 PM – 6:15 PM Resident Safety Council : “Improving Goals of Care Documentation for DNR patients” 6:15 PM – 6:30 PM Awards Stanford Quality Improvement/Patient Safety Symposium 2015 Monday, June 8, 2015 at 3:30 PM – 6:30 PM Li Ka Shing Center (LKSC), Berg Hall, LK 230A Revised 06/04/2015 Page 5 of 21 Abstracts 1. Survival and Neurological Outcomes of Cardiac Arrests at a University-Affiliated Veterans Affairs Health Care System Felipe D. Perez, MD1*;Steven K. Howard, MD2,3; Robert E. King, MS2; Edward R. Mariano, MD, MAS (Clinical Research)2,3; Geoffrey Lighthall, MD, PhD2,3; T. Kyle Harrison, MD2,3 1 Resident, Santa Clara Valley Medical Center Transitional Program; 2 Anesthesiology and Perioperative Care Service, Veterans Affairs Palo Alto Health Care System; 3 Department of Anesthesiology, Perioperative and Pain Medicine, Stanford University School of Medicine Purpose: Up to 750,000 in-hospital cardiac arrests occur annually with a small percentage surviving. There is no standardized quality metric to gauge the effectiveness of a hospital’s response to these events. Chan et al, developed the Cardiac Arrest Survival Post-Resuscitation In-hospital (CASPRI) scoring system which was used to benchmark the Veterans Affairs hospital (VA Palo Alto) resuscitation efforts. Methods: With IRB approval all patients that received chest compressions at VA Palo Alto from fiscal year 2010 to 2014 were included. CASPRI scores were used to measure the VA Palo Alto’s performance, and these results were compared to Get With The Guidelines (GWTG) national registry. Multivariate logistic regressions were performed to identify variables associated with return of a pulse and favorable neurological outcome at discharge. Results: A total of 99 patients met inclusion criterion of which 52 had return of a pulse and 30 had a favorable neurological state at discharge. VA Palo Alto was at par with the GTWG national registry (Standardized Mortality Ratio of 0.87, 95% C.I. 0.68-1.09, p= 0.26). Patients who had kidney insufficiency had an increased likelihood of having a poor neurological state at discharge (OR 5.65, 95% CI 1.40-22.75, p=0.015). Patients who had Chronic Obstructive Pulmonary Disease had an increased likelihood of not having return of a pulse (OR 9.71,95% CI 1.72-54.99, p=0.01). Conclusion: CASPRI scores can be used to measure the effectiveness or a hospital’s response to cardiac arrest and when compared to the GWTG registry it can provide a means to track quality improvement. 2. Patient-Driven Palliative Care Referrals from the Emergency Department David Wang MD*, Stephanie Harman MD Department of Emergency Medicine Purpose: To assess efficacy of a patient informational flyer in triggering referrals to palliative care from the emergency department. Methods: A one-page informational flyer about palliative care was designed with input from thought leaders in the Division of Palliative Care (PC). Patients presenting to Stanford Emergency Department (ED) who screened in for PC benefit by validated criteria were provided this informational flyer. If patients subsequently expressed interest, referrals were placed to either inpatient or outpatient PC service depending on disposition. Research protocol and just-in-time training engaged leadership among ED physicians, advanced practice providers, nursing, social work, and case management. EPIC generated orders were counted from the ED, which includes the Clinical Decision Unit. Results: In the six months prior to pilot launch (03/2014-08/2014), a total of 30 inpatient PC and 5 outpatient PC referrals were placed from the ED. In the six months following pilot launch (09/2014-02/2015), 87 inpatient PC referrals and 8 outpatient PC referrals were placed. Referrals were evenly distributed across the months. In total, the number of referrals more than doubled. Conclusion: In an increasingly resource-restrained healthcare system, EDs have an opportunity to lead in providing better care at lower costs. Early integration into palliative care has been shown to increase both quality and quantity of life while simultaneously reducing admissions and lengths of stay. A simple PC informational flyer effectively increases palliative care interest directly from the ED. Further research will explore the nuanced demographics of this captured population – underlying disease type, symptom burden, as well as satisfaction and outcomes with early referral. 3. Resident Physician Knowledge of the Discharge Medication Process and the Impact of Pharmacy Rounding Thomas Lew, MD * Department of Internal Medicine Purpose: To reduce delays in discharge due to medication issues by utilizing Transition of Care (TOC) pharmacists, as well as to assess resident knowledge of the discharge medication process. Methods: A survey sent to residents assessed the frequency of delays to discharge related to medications and knowledge of formulary information. The pharmacy rounding intervention was performed concurrently. Two Medicine wards teams were chosen for the intervention with three control teams. 
The intervention teams met with a TOC pharmacist daily to review medication lists of upcoming discharges. They identified medications that would need prior insurance authorization and alerted the teams. Control teams met with a resident to review the same information, but were not alerted to potential costly medications. Results: 155 residents were surveyed. 93% had experienced delays in discharge due to medications, many of which were due to late recognition of the necessity for prior authorizations. A significant majority of residents do not understand the Stanford insurance formulary or MediCal prescribing rules.
 In the TOC pharmacy intervention, there were four unnecessary days of hospitalization due to medications needing prior authorization, compared to zero for the intervention group. Conclusions: Delays to discharge due to medications is a common occurrence. There is a gap in knowledge among residents concerning medication coverage. Bringing in TOC pharmacists early into the discharge process, while not leading to a decrease in overall length of stay, reduced unnecessary days of hospitalization. Further utilization of TOC pharmacists would be cost-effective in reducing such delays in discharge. 4. Examining the Athlete Experience of Medical Care Calvin E Hwang, MD*; Jennifer G Baine, MD Department of Orthopedic Surgery Purpose: The medical team in the sports medicine clinic can be quite large, comprised of multiple physicians, physical therapists, athletic trainers, interns and rotating students. We hypothesized that the presence of such a large team might proclude the athletes from relaying personal information which may be crucial to patient care. Methods: We developed an IRB approved survey to better understand the athlete experience of care at the sports medicine clinic. All student-athletes presenting to sports medicine clinic over a period of two months had the Stanford Quality Improvement/Patient Safety Symposium 2015 Monday, June 8, 2015 at 3:30 PM – 6:30 PM Li Ka Shing Center (LKSC), Berg Hall, LK 230A Revised 06/04/2015 Page 6 of 21 opportunity to complete the survey. We asked questions on the level of athlete comfort in expressing concerns about their care, bringing up potentially sensitive topics, and confiding with their providers. We also collected data on the number of medical personnel in the room during the visit and demographic data. Results: Fifty surveys were collected during the study period. There did not appear to be an association between the number of providers in the clinic room and an athlete’s comfort in discussing his or her medical care or other sensitive topics. Student-athletes were comfortable asking questions during the appointment, were not afraid to express concerns, were comfortable bringing up potentially sensitive topics and asking people to leave the exam room if necessary, and did not have concerns regarding privacy. Conclusions: Although student-athletes are often seen and examined with multiple providers in a clinic exam room, they did not have any significant concerns regarding their medical care or possible violations of privacy. 5. Improving Diabetes Care and Population Health Training in an Internal Medicine Resident Clinic Anuradha Phadke MD*, Nazima Allaudeen MD, Nicole Grant MD, and Steven M. Asch MD MPH Department of Internal Medicine Purpose: Our goals were 1) improve rates of diabetic patients with optimally controlled blood pressure and annual diabetes labs and 2) improve resident exposure to population health via increased engagement with a diabetic registry. Methods: The project took place at the Palo Alto VA internal medicine teaching clinics. Initial steps included registry aggregation to determine the current state of diabetes care and review of ACGME program survey data to understand educational gaps. Subsequent steps included meetings with residents and clinic leadership, chart review, and discussion with patients to understand root causes for identified clinical gaps and to develop plan- do-study-act interventions. We then used a quasi-experimental design, with one resident clinic serving as intervention and the other comparison, to test a series of interventions. Results: We tested educational, checklist, and audit feedback/academic detailing interventions. Over a 6 month period, the percent of patients with annual hemoglobin A1c did not change in the intervention group but decreased by 5% in the comparison group. A sustained increase in patients with annual microalbumin of 2% was seen in the intervention group as compared to a decrease of 3% in the comparison group. Semi-structured interviews demonstrated increase in knowledge and behaviors surrounding population health and registry management in the intervention group. Conclusions: Audit feedback and academic detailing can increase trainee engagement with population health management. Addressing challenges of limited frequency of feedback, team protected time for registry management, and leadership prioritization, as well as focusing on patient follow-up, should further clinical improvement. 6. Decreasing Rates of Inappropriate Echo Ordering Using Report- Embedded Follow-up Recommendations Anuradha Phadke MD* and Paul Heidenreich MD, MS Department of Internal Medicine Purpose: To study if including follow-up recommendation in an echocardiography report improves appropriate use of follow-up echocardiograms. Methods: We randomized 1705 echoes, for which an echo attending chose to give a follow-up recommendation, to either include or not include the recommendation in the echo report. There were 298 recommendations to perform a follow-up study within a time frame (POS) and 1407 recommendations that a follow-up was not indicated (NEG). We conducted chart review to determine the reason for follow-up study for the 1407 in the NEG group. We defined appropriate follow-up as an echo in the recommended time frame for positive recommendations or avoidance of an echo for the same indication following a negative recommendation. Results: 805 reports were randomized to include a recommendation in the echo report (703 NEG, 102 POS) and 826 randomized to not include a recommendation (705 NEG, 121 POS). The mean age of patients undergoing echocardiography was 68 +/- 12 years. Requests for echocardiography were from general medicine outpatient (33%), cardiology outpatient (27%), other outpatient (14%), and inpatient (26%). Appropriate follow-up echocardiography following a POS recommendation occurred in 57% in the recommendation-included group and 51% for the control group. Inappropriate follow-up echocardiography (NEG recommendation) occurred in 4.1% randomized to the recommendation- included group and 5.5% for the control group. Overall inappropriate follow-up decisions occurred in 8.9% of those randomized to the recommendation-included group compared to 11.9% of those randomized to control (p=0.05). Conclusion: Placing recommendations within echo reports decreased rates of inappropriate follow-up. 7. Barriers to Effective RN-MD Communication at Stanford Adam Sang* MD, Ian Chong MD, Minjoung Go MD, Ryan Perumpail MD, Arghavan Salles MD, Michael Shaheen MD, Barbara Mayer RN PhD, Ruth Fanning MD Resident Safety Council Purpose: Effective communication between nurses and physicians is essential to ensure optimal patient care, foster a culture of safety, and improve job satisfaction. Yet, both qualitative and quantitative data, including work experience, National Database of Nursing Quality Indicators (NDNQI) survey data, and literature review, suggest that RN-MD communication is suboptimal across many major academic centers. Our goal as a multi-specialty committee of residents is to start the process of improving RN-MD communication at our institution by gathering data on specific barriers to effective communication from both professions. Methods: Communication barriers were explored by focus groups comprised of physicians and nurses across many disciplines. Nursing NDNQI data from our institution was reviewed from the past academic year. Result: Resident physicians considered the following as barriers to effective communication with RNs: 1) lack of understand of each other’s roles, 2) technological limitations, 3) conflicting personalities, and 4) the intrinsic stressful and fast-paced nature of health care. From the nurses’ perspective, the barriers include 1) lack of face time between RNs and MDs, 2) lack of understanding of each other’s roles, 3) poor closed-loop communication, 4) cultural and professionalism conflicts, and 5) lack of good feedback mechanisms. Conclusions: Institutionally, there is much interest from both RNs and MDs to improve the effectiveness of communication between both professions. A number of common barriers were identified by both physicians and nurses, including a lack of understanding of each other’s roles, and culture/personal conflicts. The overall goal of this project is to tailor interventions which address these specific barriers to communication, such Stanford Quality Improvement/Patient Safety Symposium 2015 Monday, June 8, 2015 at 3:30 PM – 6:30 PM Li Ka Shing Center (LKSC), Berg Hall, LK 230A Revised 06/04/2015 Page 7 of 21 as education pieces, multi-disciplinary training, and institutional support to increase face time. 8. An iPad Application for Assessing Risk of Complication Occurrence in Spine Surgery Anand Veeravagu*, Christian Swinney, Amy Li, Adrienne Moraff, John Ratliff Department of Neurosurgery Purpose: The ability to predict risk of perioperative complications based on patient and procedure specific factors would benefit surgeons and patients. Current methods of risk assessment for spine surgery are limited. This study assesses an iPad application that predicts the risk of perioperative complications based on patient and procedure specific unique factors. Methods: We developed a smooth data entry process using an iPad device that provided appropriate inputs into a previously developed computational model of adverse event occurrence in spine surgery. The model was previously generated using longitudinal prospective data, consisting of 279,145 records. The present study applied this model to a group of 200 patients. Patient factors were entered into the application's interface. Predicted complications were then compared to the actual complications for the 30 day postoperative period. Results: The mean predicted probability of experiencing a complication was .4494 for patients experiencing 1+ complications and .3714 for patients experiencing 0 complications (p=.0436), according to the iPad application. In the current study a group of 70 patients (35%) experienced post- operative complications. The most common comorbidity in patients experiencing complications was hypertension (49%, compared to 36% in patients with no complications). Conclusion: The iPad application and the statistical model that it utilizes can provide the surgeon with a straightforward and timely means of assessing the risk of perioperative complications based on known patient factors and comorbidities. This device has the potential to improve both patient safety and the quality of medical care. 9. Urologic Guidelines for Evaluation of Patients Presenting with Suspected Urolithiasis at Stanford Emergency Department Dimitar Zlatev MD*, Simon Conti MD, Remy Lamberts MD, Phil Harter MD, Harcharan Gill MD, Eila Skinner MD Department of Urologic Surgery Purpose: Pain from urolithiasis is a common presentation to the emergency department (ED) in the United States. Abdominal non-contrast computed tomography (NCCT) has been the standard initial imaging method for these patients. The use of abdominal ultrasonography (US) remains an area of ongoing research, with lower cumulative radiation exposure but lack of definitive localization and sizing of stones necessary for surgical intervention. We sought to create guidelines for evaluation and urologic referral of patients presenting with suspected urolithiasis to Stanford Hospital & Clinics ED. Methods: A comprehensive literature review was performed, including MEDLINE searches and the most recent guidelines from the American Urological Association and European Association of Urology. This analysis formed the evidentiary basis of the recommended guidelines, with additional attention dedicated to streamlined ED workflow and optimized department-specific urologic management. Results: Initial evaluation should include a BMP, CBC, urinalysis, urine culture, and plain abdominal radiography. Abdominal US is optional. Medical expulsion therapy with outpatient referral to Urology is appropriate for afebrile patients with unilateral urolithiasis, satisfactory pain control, and no evidence of sepsis or renal failure. NCCT (low-dose if BMI < 30) should be obtained for all patients with evidence of sepsis, renal failure, uncontrolled pain, or non-diagnostic initial imaging. Inpatient consultation to Urology is required for all patients with evidence of sepsis, renal failure, solitary kidney, bilateral obstructing urolithiasis, or uncontrolled pain. Management should otherwise include medical expulsion therapy and outpatient referral to Urology. Conclusions: The proposed urologic guidelines have the potential for streamlining initial patient evaluation of suspected urolithiasis, with the goal of enhanced emergency department workflow and optimized urologic management. 10. Standardization of Hospital Discharge Instructions Templates following Admission to Urologic Surgery Service Dimitar Zlatev*, Kerri Stevenson, Remy Lamberts, David Guo, Harcharan Gill, Eila Skinner Department of Urologic Surgery Purpose: The transition of care from hospital to home is a vulnerable time for patients. Discharge instructions that are accurate, clear, and complete can reduce complications and readmissions, while ineffective instructions can adversely affect patient care. There remains significant variability in the content of discharge instructions across providers in our department. We sought to create standardized, diagnosis-specific electronic templates to be used for patients discharged from the Adult Urologic Surgery service at Stanford Hospital & Clinics. Methods: Electronic discharge instructions were created as diagnosis- specific SmartPhrases for use within the Discharge Navigator in EPIC. Each template included the following standard components: name of procedure or reason for admission, diet, activity, wound care, drain care, medications, follow-up, warning signs, medical contact information during normal working hours (including clinic telephone numbers), and medical contact information after normal working hours. EPIC wildcard placeholders (denoted as ***) were utilized to allow incorporation of patient-specific details. Standardized templates were created for the most common urologic operations and personalized for each of the 9 attending physicians within the Urologic Surgery department. Results: A total of 79 diagnosis-specific discharge instructions were created. The average number of discharge instructions was 9 (range 3 to 16) per attending physician. All discharge instructions templates were shared in EPIC with urology residents and nurse practitioners for clinical use. Ownership was limited to select users to maintain template integrity. Conclusions: Standardized diagnosis- and attending-specific discharge instructions templates have the potential of easing the transition of care following hospital discharge, with the goal of improved patient satisfaction and reduced outpatient complications and readmissions. 11. Design of emergency cognitive aids impacts performance on information retrieval and use: A simulation-based study in physicians Janak Chandrasoma, MD, Educational Informatics Fellow, Stanford AIM Lab Anna Clemenson, RN, Stanford AIM Lab Reuben Eng, MD, Educational Informatics Fellow, Stanford AIM Lab Kyle Harrison, MD, Core Faculty in Simulation, Stanford AIM Lab Larry Chu Stanford AIM Lab, Anesthesiology, Perioperative and Pain Medicine Purpose: To determine the effect of pictographic design elements on human factors performance of ACLS crisis management, compared to text- based aids. Stanford Quality Improvement/Patient Safety Symposium 2015 Monday, June 8, 2015 at 3:30 PM – 6:30 PM Li Ka Shing Center (LKSC), Berg Hall, LK 230A Revised 06/04/2015 Page 10 of 21 dataset was narrowed to the core rotations during intern year of one residency class. Results: A total of 20,280 notes and 112,214 orders were written by 26 pediatric interns between June 2012 and June 2013. The average number of notes written was 780 (range 595 to 945) with a standard deviation of 80. The average number of orders was 4,316 (range 1,776 to 6,004) with a standard deviation of 1,057. On average, the black cloud residents – defined as those with number of notes and orders in the top quartile – wrote 19% more notes and 91% more orders than the “white cloud” residents in the bottom quartile. Conclusions: There exist significant variations in workload between pediatric residents at Stanford Children’s Health. Disproportionate resident workload may contribute to heterogeneous opportunities for resident education, physician wellness, and quality of patient care. 18. Assessment of Anti-Infective Utilization at Stanford Hospital Maria Birukova, Elena Mancini, Sam Roosz, Stefanie Seisenberger Glenn, Aslihan Selimbeyoglu, Brannon Weeks, Elizabeth Zambricki** Division of Infectious Diseases, Department of Medicine Objective: In 2014, Stanford General Medicine ranked 78th of 98 University Health Care (UHC) hospitals in antimicrobial costs per patient, with costs almost 3x the mean cost per patient for UHC hospitals. Our objective was to analyze UHC data to identify areas and anti-infectives that represent opportunities for savings and compare Stanford anit-infective utilization to that of peer hospitals with equivalent outcomes and lower costs. Study Design: Quantitative Data analysis comparing Stanford Antibiotic use to other peer UHC health systems Methods: Top spending Stanford Diagnosis Related Groups (DRGs) were identified and selected based on the total potential cost savings. Total cost savings was calculated by estimating the Stanford mean cost per case minus the UHC mean cost/case times the number cases at Stanford. After uncovering the top 12 DRGs, physicians within that diagnosis group were interviewed to get a better understanding of the practices in high-spending groups. Results: The 12 DRGs prioritized for analysis collectively represent an opportunity for $1.6M in annual cost savings compared to the UHC mean. Top DRGs included acute leukemia, ECMO, heart transplant, otitis media and URI, infectious and parasitic disease, liver transplant, ventilator support, allogeneic BMT, liver disorders, autologous BMT, pancreas/liver procedures and upper GI procedures. Anti-infectives that represented the most high cost divergence from other UHC systems were caspofungin, ribavirin, amphotericin B. Conclusion: Utilization of anti-fungals and ribavirin are significant outliers at Stanford relative to peer hospitals. Additionally, Stanford prescribes each prioritized agent in a significantly greater proportion of cases than does its peers. 19. Improving the Quality of Mammographic Positioning to Detect Breast Cancer Christina Chen*, Audrey Strain, Jake Mickelsen, David Larson, Robert Jesinger, Daniele Botelho, Sandra Fromholz, Chrystal Obi, Alexis Crawley, Jafi Lipson, Debra Ikeda, Claudia Cooper, Sunita Pal Breast Imaging and Mammography at Stanford Hospital Purpose: High quality screening mammograms are essential for radiologists to detect breast cancer. The American College of Radiology (ACR) outlines 13 criteria of breast positioning to optimize mammograms. Our goal is to increase the percentage of Stanford screening mammograms achieving the ACR criteria from a mean of 65% in 2013 to 90% by June 2015. Methods: The Radiology Improvement Team Education course sponsored this project. Team members identified causes that barred achieving the ACR criteria, such as disagreement on what met ACR criteria, not having a standard work to acquire and read mammograms, and lack of communication between technologists and radiologists. From the time period of July 2014 to March 2015, over 1,700 mammograms were audited. Results: Developments to address the causes included: teaching modules on what meets ACR criteria, standard work for radiologists to recall mammograms that did not meet ACR criteria, system for the technologist to document why criteria were missed, auditing system to track performance, and feedback sessions between technologists and radiologists. By March 2015, 10 of the 13 ACR criteria were being met by the target goal of 90% of mammograms. Conclusion: The most recent quality improvement publication on breast positioning dates to 1993. To our knowledge, Stanford is the first breast imaging department to proceed in a structured way to meet the ACR positioning criteria. Sustainment plans include hiring a mentor to provide real-time feedback, scheduling sessions of radiologists and technologists to review ACR criteria, and automating feedback to technologists. 20. The Impact of Health Outcomes Information Registries on Future Graduate Medical Education Neil Ray MD*, Jordan Newmark, MD, Ming-Chih Kao, MD, Sean Mackey, MD Division of Pain Medicine, Department of Anesthesiology Purpose: The improvement of graduate medical education and patient quality of care remains a constant and continual goal. As large health data registries develop, one new method of advancing both goals involves using patient outcomes information. Our goals are to describe the development and design of our collaborative health outcomes information registry (CHOIR) and determine whether it allows for trainees to meet high-level 4 and 5 ACGME patient safety and quality improvement competencies. Methods: We anonymously surveyed existing trainees on whether they feel the use CHOIR, allows them to meet high-level 4 and 5 ACGME patient safety and quality improvement competency goals. Level 4 and 5 milestones being that trainees identify opportunities in the care of patients to improve patient outcomes and define outcome measures leading to quality improvement projects. Results: CHOIR is a robust computer adaptive testing survey system that is used to electronically survey and track patients on valuable clinically relevant pain-related outcomes such as depression, anxiety, anger, etc. The system also assesses and computes scores to standardize the biopsychosocial data collected. In addition, aggregate data is used to study overall trends and associations. We found, trainees agreed or strongly agreed that CHOIR increases patient quality of care, decision- making, and supports aggregate research efforts. Conclusion: Simply using CHOIR enables trainees to meet high-level 4 and 5 ACGME patient safety and quality improvement competency goals. Increased use of health registries, like ours, will play an integral role in improving both future graduate medical education and patient quality of care. 21. Obtaining complete, pertinent clinical histories for radiographic examinations, a pilot project. Benjamin Johnson*, Angela Fast, Anita Angelotti, Jake Mickelsen, Geoffrey Riley, David Larson Department of Radiology Stanford Quality Improvement/Patient Safety Symposium 2015 Monday, June 8, 2015 at 3:30 PM – 6:30 PM Li Ka Shing Center (LKSC), Berg Hall, LK 230A Revised 06/04/2015 Page 11 of 21 Purpose: To improve the completeness and quality of the clinical histories which provide the context in which radiologists interpret radiographic studies. Methods: From November 2014 through March 2015, this project was piloted at the Blake Wilbur outpatient site. Prior to implentation a team of radiologists and technologists defined what constituted an adequate and appropriate clinical history. This definition includes “what” an order provider is looking for, “when” the symptoms began, and “where” specifically are the symptoms located. Prior to implementation, technologists at this site were educated about appropriate clinical histories and briefly trained on how to obtain them prior to completing an examination. In the months prior to and following implementation, radiology residents audited both the ordering provider clinical history and technologist obtained clinical history for each examination peformed at the Blake Wilbur site. Results: At the Blake Wilbur outpatient radiography site, complete clinical histories were only present for 5-15% of studies audited in the month immediately preceding implentation of the quality improvement project described above. Upon implementation of technologist acquired clinical histories during examinations, this improved gradually over time, and for the last 4 weeks of the project ranged from 92-98%. Conclusions: Appropriate and complete clinical histories provide key context for radiologic interpretation of radiographic studies. Unfortunately, due to multiple factors, more often than not radiologists are not provided with adeuqate clinical histories from ordering providers. As a department, we are able to significantly improve the quality and consistency of clinical histories by obtaining them directly from patients at the time of examination. 22. Root Cause Analysis of Internal Medicine Hospital Readmissions within 30 Days Andrea Smeraglio, MD1*; Gomathi Krishnan PhD2; Jonathan Chen, MD1, Daniel Croymans3, Trishna Narula4, Lisa Sheih, MD1 1 Department of Internal Medicine, Stanford University School of Medicine, Stanford, California, USA 2 Clinical and Translational Research Department of Internal Medicine Purpose: This project focused on determining the underlying systems issues that lead to hospital readmissions from the internal medicine service at Stanford Hospital. We compiled information from chart reviews and patient based interviews to fully elucidate the reasons behind hospital readmissions. Methods: We used STRIDE data in combination with manual chart reviews to evaluate risk factors and predictors for hospital readmission within 30 days after discharge from the Stanford Internal Medicine service. Accuracy of STRIDE data was confirmed using random chart biopsy. Patient interviews were conducted over the phone by medical students within 3 months of hospital discharge. Results: Preliminary results based on the initial 15 manual chart biopsies and patient interviews indicate that risk factors for hospital readmission included multiple medical comorbidities, complex social situations, lack of RN education at time of discharge, lack of discharge follow-up call by RN, lack of consult to discharge pharmacy. Conclusions: Multiple avenues can be addressed to decrease hospital readmissions on the internal medicine service. Further data collection is planned to obtain a larger sample with the addition of STRIDE data for a larger sample size. 23. Building a Clinical Effectiveness Program at LPCH: Establishing a Prototype Clinical Pathway Claudia A. Algaze*, Catherine D. Krawczeski, Andrew Y. Shin, Chealsea Nather, Chandra Ramamoorthy, 
Komal Kamra, Alaina Kipps, Vamsi Yarlagadda, Marcy Lamonica, Monica Mafla, Krisa Elgin, Juanita Hickman, 
 Tanushree Vashist, Paul Sharek Department: Pediatrics, Division: Cardiology and the Center for Quality and Clinical Effectiveness at LPCH Purpose: Clinical pathways, by reducing unnecessary variation, can improve patient outcomes and efficiency. We lack a reliable method to develop, implement, monitor, and improve clinical pathways. Without this methodology, our impact and sustainability is limited. By implementing and testing a prototype clinical pathway, we are identifying factors associated with a successful Clinical Effectiveness (CE) program and using these lessons to deploy a hospital-wide CE program at LPCH Stanford. Methods: The model includes selection of a prototype population, building of a multidisciplinary team, clinical pathway build and deployment, development of analytic support and scalability. The postoperative management of tetralogy of Fallot was selected as the area of clinical focus. An evidence-based clinical pathway was developed from rigorous review of current literature and best practices. We widely socialized the pathway to practitioners, nurses, and other allied health workers. As we conduct a paper-based pilot, we are concurrently developing implementation tools (order sets, electronic medical record-based decision support tools, and monitoring tools for real-time data extraction and analysis to evaluate our improvement through process and outcomes metrics. Primary outcome measures include postoperative length of stay (LOS), cardiovascular intensive care unit (CVICU) LOS, duration of mechanical ventilation and resource utilization. Secondary outcome measures include patient/family and provider satisfaction. Process measures include compliance to pathway enrollment and completion. Balancing measures include incidence of reintubation, readmission to CVICU and related readmission to hospital. Results: There has been a reduction in LOS, CVICU and duration of mechanical ventilation after implementation of the TOF clinical pathway. 24. Pediatric Anesthesia Critical Incident Reporting: Understanding Motivators and Barriers to Anesthesiologists’ Use of an Intra-Operative Self-Reporting System *Matthew Muffly, Tom Caruso, Jumbo Williams Department of Anesthesia, Perioperative and Pain Medicine; Division of Pediatric Anesthesia Background: Obtaining reliable data related to critical incidents is important in identifying and planning quality and safety initiatives. A critical incident reporting tool is embedded in our intraoperative charting system; however, attending physician frequency of reporting by percentage of cases varies (0-12%). To understand this variation, we characterized the motivators and barriers to reporting anesthesia-related critical incidents. Methods: We conducted interviews with pediatric anesthesiology faculty to describe motivators and barriers to self-reporting. We then tested respondents’ knowledge of reporting definitions and assessed variation in reporting thresholds with three clinical vignettes. Results: Twenty-eight (76%) faculty participated, the majority (74%) of whom agreed that reporting was useful, yet most (59%) agreed that more training was needed to clarify reporting definitions. Only 28% correctly identified all three clinical vignettes as reportable in accordance with established definitions; of those, only three (11%) would actually report all three incidents within the current system. Stanford Quality Improvement/Patient Safety Symposium 2015 Monday, June 8, 2015 at 3:30 PM – 6:30 PM Li Ka Shing Center (LKSC), Berg Hall, LK 230A Revised 06/04/2015 Page 12 of 21 Motivators to reporting included: the ability to identify systems issues, improve patient safety, increase faculty education, and duty. Barriers included: lack of sufficient feedback, concern about punitive repercussions, forgetfulness, and variation in individual reporting thresholds. Conclusion: These semi-structured interviews allowed us to identify motivators and barriers to anesthesiologist critical-incident self-reporting. While the majority of anesthesiologists agree that the system is useful, barriers to reporting critical incidents exist and may account for reporting variation. We plan to address these barriers through a combination of educational initiatives, improved feedback and increased transparency related to the critical incident reporting process. 25. Alendronate as treatment for Osteonecrosis of the Femoral Head in Sickle Cell Disease Oyebimpe Adesina*, Michael Jeng, Carolyn Hoppe and Jason Gotlib Divisions of Hematology and Oncology, Department of Medicine Purpose: Sickle cell disease (SCD) is the most common disorder of hemoglobin in the world, affecting up to 100,000 people in the United States, and millions more in developing countries. SCD is the most common cause of osteonecrosis of the femoral head (ONFH) in childhood, and affects up to 50% of adults with homozygous SCD. There is no standard of care for the treatment of SCD-related ONFH. Bisphosphonates have shown efficacy in the management of steroid-induced and idiopathic ONFH, but have never been studied in SCD patients with ONFH. Methods: We propose a randomized, double-blinded, placebo-controlled prospective clinical trial of alendronate in SCD patients, ages ≥ 12 years, with radiographic evidence of early stage ONFH. We will also measure levels of plasma microparticles (MP), membrane-bound vesicles, which have been implicated in the pathogenesis of SCD-related ONFH. Results: The primary endpoint is a ≥15-point improvement on the Children’s Hospital Oakland Hip Evaluation Scale (CHOHES) score, a validated tool that measures hip pain and function in SCD patients with ONFH. Preliminary studies from Children’s Hospital in Oakland showed statistically significantly higher total MP levels in a cohort of SCD patients with ONFH, compared to age-matched SCD controls. We hypothesize that MP levels can predict early onset ONFH in SCD patients, and can be used to monitor response to bisphosphonate therapy. Conclusion: Advances in early diagnosis and effective non-surgical treatment of ONFH in SCD will address significant unmet clinical needs in this vulnerable patient population, and help to improve their quality of life. 26. Improving Goals of Care Documentation for DNR patients Molly Kantor*, Lance Downing, David Wang, Daniel Fang, Robert Fairchild, Yi-Ren Chen, Teng Lu, Sara Stern-Nezar, Stephanie Harman, Stephanie Bowen, Paul Maggio Resident Safety Council: Internal medicine, Neurosurgery, Emergency Medicine, Neurology, General surgery Purpose: We aimed to improve communication of goals of care for inpatients. Only 41% of inpatients with do-not-resuscitate (DNR) code status, including all patients with code status orders of "do-not- resuscitate/do-not-intubate," "do-not-resuscitate/comfort measures only," "do-not-resuscitate/do-not-escalate," and "partial code," had goals of care documented in designated "Goals of Care" (GOC) notes. We sought to improve rates of use of GOC notes for patients with DNR code statuses to 80%. Methods/Results: To better understand the problem, we performed standard interviews of residents in internal medicine, neurology, neurosurgery, hematology, oncology, and medical, surgical, and cardiac intensive care. From this, we created a root cause analysis and identified a that a key root cause was lack of standard work, including lack of clarity about responsibility for writing a note, forgetting to write the note, and finding the GOC note template too cumbersome. We then developed standard work that identified the team resident as responsible for ensuring a GOC note was written for all patients with DNR code statuses, with the attending physician responsible for co-signing these notes. We simplified the GOC note template. We are in the process of disseminating these changes in standard work via an educational campaign to residents. Finally, we plan to use the electronic medical record to help residents remember to place goals of care notes through a Best Practice Alert. This will be a non-interruptive reminder that appears if a patient has a DNR code status but no GOC note for the current inpatient encounter; this is still being finalized. After this is in place, we will monitor whether rates of GOC notes for DNR patients improve to our set goal. Conclusions: As frontline workers, residents are a key resource for identifying quality improvement problems and their underlying causes. Many patients with DNR code statuses lack GOC notes, and we aim to improve this via setting standard work. 27. Improving Pathology Curriculum for Nephrology Fellows Enrica Fung, MD, MPH*; Orlando Camacho, MD; Michelle O’Shaughnessy, MB BCh MRCPI; Ade A. Taiwo, MD; M. Gabriela Velez, MD, PhD., Timothy W. Meyer, MD, Neeraja Kambham, MD. Division of Nephrology, Department of Medicine Purpose: Fellows were previously exposed to renal pathology through attendance at monthly hour-long “biopsy conferences” where faculty and fellows discussed difficult cases; and informal quarterly “fellows-only conferences” where fellows asked pathologists questions on core topics. Our project sought to improve fellows’ education in renal pathology. Methods: Two renal pathologists led an orderly review on the following topics, dedicated to fellows only: 1. Common renal pathology; 2. Primary glomerulonephritis; 3. Secondary glomerulonephritis; 4. Plasma cell dyscracia-related renal disease; 5. Renal transplant pathology. Second year fellows helped in preparation of teaching material and fellows were encouraged to ask questions. Participants completed a survey measuring their level of satisfaction with renal pathology teaching and their confidence in interpreting renal biopsy findings before and after completion of this teaching series. Five 10- question pre-session quizzes and a comprehensive 50-question post- session final measured knowledge acquisition. Results: Eleven of twelve 1st–3rd year fellows responded to the pre-course survey. Five fellows felt “somewhat dissatisfied” with the current training, while six felt “somewhat satisfied” or “satisfied” with the current training. Ten respondents reported being “not confident at all” or “somewhat lacking in confidence” in their abilities to interpret biopsy findings, while one reported being “somewhat confident”, prior to our project. To date, fellows answered 63% questions correctly on the knowledge-based pre- session quizzes. Conclusions: Survey of nephrology fellows suggested poor confidence in interpreting renal pathology despite perceived importance of the topic. Survey and testing to be completed May 2015 would reveal whether our project improved confidence and knowledge. 28. Covert Observations, Auditing, and Videos: A Multi-faceted Approach to Improving Hand Hygiene Adherence at LPCH. Kaitlyn Phuong Le, MD*, Lauren Destino, MD, Dianne Laumann, RN, MBA, PMP, Terry Platchek, MD, and Amit Singh, MD. Stanford Quality Improvement/Patient Safety Symposium 2015 Monday, June 8, 2015 at 3:30 PM – 6:30 PM Li Ka Shing Center (LKSC), Berg Hall, LK 230A Revised 06/04/2015 Page 15 of 21 Purpose: A supportive working environment has been shown to be a protective factor against the corrosive effects of residency.However, there are few published interventions on how to foster this supportive environment. The Golden Ticket Project (GTP) is a novel wellness pilot program that encourages peer recognition within a pediatric residency program. The purpose of this pilot was two-fold: (1) To evaluate the feasibility of the GTP and (2) To characterize the qualities residents appreciate about each other across all levels of residency training, as revealed through participation in the GTP. Methods: All residents were eligible to participate in the GTP. Participating residents nominate a co-resident for any behavior they perceived as going “above and beyond”. Nominations could be made through an online form during the pilot period (October 2013-July 2014). Golden ticket recipients were offered a token prize every month, and the reasons for the nominations were posted near the residency lounge. As part of an IRB- exempt study, two reviewers independently analyzed each nomination to code for emerging themes; a consensus on core themes was reached. In August 2014, an anonymous Likert survey was distributed to all residents to assess their attitudes about the GTP. Results: A total of 50/83 (60%) residents gave or received golden tickets. 56 tickets were given representing 63 codes) during the 10-month pilot period. Golden ticket nominations represented five major themes: Positive Attitude, Teamwork, Patient Care, Resident-Resident Support, and Strong Supervisory Skills. Across all classes, the most recognized theme was Teamwork, noted in 44% of awarded tickets. PGY2 and PGY3s gave golden tickets proportionately more for Patient Care than PGY1s. According to the residents who participated in the post-pilot survey, residents on average agreed that because of the GTP, they were “more aware of acts of kindness in the residency program.” 76% of survey participants indicated they wanted the program to continue. Conclusion: The GTP is a feasible wellness program, valued by participating residents. The GTP offers a needed venue for residents to recognize teamwork and other positive attributes of their co-residents. The program can be implemented at other institutions and may serve as a framework for robust wellness interventions. 37. Utility of Focused Re-screening of Liquid-based Papanicolaou Tests Diagnosed as Negative for Intraepithelial Lesion/Malignancy with Positive High-Risk HPV in an Academic Tertiary Care Center: Comparison with Follow-up Cytology/Biopsy Adam J. Gomez*, MD, Thuy Penedo, MD, Tala Lo-Guyamatayo, CT, Harumi Lim, CT, Steven R. Long, MD, Christina S. Kong, MD Division of Cytopathology, Department of Pathology Purpose: Recent studies have supported focused re-screening of Pap tests that are interpreted as negative for intraepithelial lesion or malignancy (NILM) but positive for high-risk HPV DNA by co-testing. While the percentage of women who fall into this category is reported to range from 3.6% to 5.4%, the increased risk of squamous dysplasia in these patients may impact screening guidelines and triage to colposcopy. In this study, we examine the impact of focused re-screening for this population at our institution. Methods: A search for Pap tests with HPV co-testing from January to March 2013 yielded 753 cases, and 45 (5.97%) were NILM/HPV+. 44/45 of these cases were availble for re-screening by two cytotechnologists, with further review by two board certified cytopathologists. Follow-up Pap test and biopsy findings were obtained from the pathology information system. Results: 37/44 (84.1%) cases were interpreted as NILM on re-screen, and 1 of these had evidence of dysplasia (HSIL) on follow-up biopsy. 7/44 (15.9%) cases were upgraded to atypical (ASCUS or ASC-H) on re-screen, and 3 of these had evidence of dysplasia (2 LSIL, 1 HSIL) on follow-up Pap or biopsy. Conclusions: HPV screening is more sensitive than Pap, resulting in the detection of HSIL in a patient with a confirmed negative Pap test (1/44). Re- screening of NILM/HPV+ Pap tests could have led to earlier detection of HSIL in 2.3% (1/44) of cases. Inherent bias from re-screening HPV+/negative Pap tests may lead to overcalls of atypical squamous cells (ASCUS or ASC- H). 38. Stress Ulcer Prophylaxis: Clinical Guideline Implementation in the Intensive Care Unit Cody A. Parsons, PharmD*, Critical Care Resident; Hangyul Chung-Esaki, MD, Critical Care Fellow; Nicholas Berte, RN, BSN, Critical Care Nurse Department of Pharmacy Purpose: Stress ulcer prophylaxis (SUP) decreases the incidence of gastrointestinal bleeding (GIB) in patients receiving mechanical ventilation or those with coagulopathy.1 However, its wide-spread use is associated with increased rates of Clostridium difficile colitis infections and nosocomial pneumonia without decreased GIB rates in low risk patients. 2,3,4,5 A survey of the intensive care units (ICUs) at Stanford in 2012 by Wong et al revealed that approximately 28% of patients inappropriately received SUP. By establishing an evidence-based clinical practice guideline with real-time feedback to clinicians, we seek to reduce the rate of inappropriate SUP in ICU patients, and consequently reduce the risk of C. difficile infection and nosocomial pneumonia. Methods: We conducted a review of the literature review to establish an evidence based clinical practice guideline in conjunction with the ASHP national guidelines, and created a Pharmacy-driven protocol to monitor and regulate the use of SUP in the ICU. Per the protocol, Pharmacy staff will prospectively evaluate active orders on every Monday, Wednesday, and Friday and discontinue inappropriate SUP orders. The protocol was approved by the ICU CQI committee, and implemented on April 1, 2015 after initial education and dissemination of the guidelines to clinicians. Cumulative rates of inappropriate SUP, as well as secondary outcomes such as rates of GIB, C. difficile infection, and nosocomial pneumonia will be monitored from April through June 2015 and compared with baseline data from April through June 2014. Results and Conclusions: This project is currently active with pending data. References: 1. ASHP Commission on Therapeutics. ASHP therapeutic guidelines on stress ulcer prophylaxis. Am J Health-Syst Pharm. 1999; 56:347-79. 2. Buendgens L, Bruensing J, Matths M, et al. Administration of proton pump inhibitors in critically ill medical patients is associated with increased risk of developing Clostridium difficile-associated diarrhea. J Crit Care. 2014 Aug;29(4):696.e11-5. 3. Eom CS, Jeon CY, Lim JW, et al. Use of acid-suppressive drugs on risk of pneumonia: a systematic review and meta-analysis. CMAJ. 2011;183:310- 319. 4. Howell MD, Novak V, Grgurich P, et al. Iatrogenic gastric acid suppression and the risk of nosocomial Clostridium difficile infection. Arch Intern Med 2010; 170:784-790. 5. Miano T, Reichert M, Houle T, et al. Nosocomial pneumonia risk and stress ulcer prophylaxis: a comparison of pantoprazole vs ranitidine in cardiothoracic surgery patients. Chest. 2009 Aug;136(2):440-7. Stanford Quality Improvement/Patient Safety Symposium 2015 Monday, June 8, 2015 at 3:30 PM – 6:30 PM Li Ka Shing Center (LKSC), Berg Hall, LK 230A Revised 06/04/2015 Page 16 of 21 39. Development of a Standardized Pre-anesthesia Pregnancy Screening Protocol Erin Crawford*, Chris Clave, Lindsay Borg, Jessie Ansari, Lindsey Bergman, Victoria Fahrenbach, Tyler King, Thomas Caruso Department of Anesthesiology, Perioperative and Pain Medicine Purpose: Through the development of a standardized pregnancy screening protocol, we aim to eliminate administration of anesthesia to patients with an undetected pregnancy. Methods: Design The study consists of the following three phases:  Analysis of Current State (April 15 – June 1, 2015): Identification of childbearing patients receiving pre-anesthesia pregnancy screening.  Education and Implementation (June 1 – July 15, 2015): Gap analysis followed by education of staff of new process to optimize pregnancy screening.  Post Intervention audit (July 15 – September 1, 2015): Review intervention compliance and examine percentage of patients who receive appropriate screening. Study Population Randomly selected medical records of 210 females ages 11-55 having surgery in the Stanford Main Operating Rooms or Ambulatory Surgery Center were searched for a serum or urine pregnancy test on day of surgery or within one week prior. Patients were exempt from testing if the patient was offered and declined a pregnancy test, last menstrual period occurred within 21 days prior to surgery, or the patient had a history of hysterectomy, bilateral oophorectomy, or bilateral tubal ligation. Data Analysis: Two sample t-tests will be used to examine the differences in the percentage of women who receive pregnancy screening after the intervention. Results: Data collection and analysis is ongoing. Preliminary data reveals a significant proportion of female patients of childbearing age not receiving a pregnancy test prior to surgery, without a documented exemption. Conclusions: This task force will work to develop a standardized approach to pregnancy testing to improve compliance with nationally accepted standards of care. 40. Reducing functional MRI scan times by optimizing workflow Wilson B. Chwang*, Michael Iv, Darryl Costales, Jason Smith, Teresa Nelson, Aleksandrs Kalnins, Jake Mickelsen, Roland Bammer, Dominik Fleischmann, David B. Larson, Max Wintermark, and Michael Zeineh Department of Radiology Purpose: Functional MRI (fMRI) is a specialized examination requiring coordinated efforts of the entire health care team. We observed that in our practice, fMRI scan times were lengthy. Our purpose was to reduce fMRI scan times by increasing workflow efficiency. Methods: We reviewed all fMRI exams performed from January 2013 to April 2015. We plotted the scan times on a run chart, and performed root- cause analysis. We identified key drivers, and specific interventions which were 1) eliminating intravenous contrast, 2) reducing repeated language paradigms, 3) updating technologist checklists for patient monitoring, 4) updating visual slides and audio, 5) developing a patient training video, and 6) developing multilingual paradigms. Results: We performed 89 fMRI exams from January 2013 to April 2015. One exam was excluded since the patient was unable to complete the study. For the remaining 88 exams, the mean scan time was 73 minutes, median was 70 minutes, and range was 27 to 148 minutes. We implemented four specific interventions. The outcomes data met criteria to indicate a shift in the process median on November 28, 2014. Prior to this date, there were 72 fMRI exams with a median scan time of 74 minutes. After this date, there were 16 fMRI exams with a median scan time of 59 minutes. Conclusion: By implementing specific interventions to improve our workflow, we successfully reduced our median fMRI scan times from 74 to 59 minutes. We believe that our process of workflow optimization can be applied broadly to any functional MRI practice. 41. Improving Face-to-Face Handoff in Pediatric Intensive Care to Acute Care Transfer Michael Tchou*, Mihaela Damian, Lauren Destino Department of Pediatrics Purpose: Residents identified that verbal handoff from pediatric intensive care unit (PICU) to acute care ward (ACW) providers was often incomplete or delayed. Providers were unable to evaluate patient for appropriateness of transfer. Our project aim was to implement a face-to-face handoff process within one hour of PICU to ACW transfer. Methods: Residents worked with key stake-holders to map the current transfer process and develop a new process. The new process used unit clerks, EHR orders, and nurse-to-MD notification of the transfer. This process begun on June 2014 and a self-reported audit of PICU to ACW transfers was used to track outcomes. Results: After the first transfer process improvement, only 24% of handoffs were face-to-face. After implementation of the second process, within 3 months, face-to-face handoffs increased to >80% and >60% of the time residents were paged prior to patient arrival on the floor and patients arrived within one hour of handoff. Conclusions: With resident engagement, education and a simple transfer process, rapid improvement was seen over 3 months. Maintaining this process proved difficult with declining participation in self-report surveys, which may have skewed results towards reporting only transfers with non- adherence to standard process. Declines in patient arrival within one hour appear to be related to a true change in adherence to the process requiring future rounds of improvement. The aim of 100% face to face communication continues to be improved with the help of residents and attending leads. 42. Outcomes from an Advanced Course in Quality Improvement and Performance Improvement Michael Tchou, MD*; Nivedita Srinivas, MD; Terry Platchek, MD; Alyssa Bogetz, MSW; Rebecca Blankenburg, MD; Lauren Destino, MD Department of Pediatrics Goals and Objectives: The Quality Improvement and Performance Improvement (QI/PI) foundations course is part of a longitudinal track during residency which was developed to provide advanced training above ACGME required QI/PI education and activities for residents considering careers in these fields. The goals of the course were to provide residents: 1) advanced leadership training in Lean operations management and improvement science; 2) hands on experience in QI/PI methodology; 3) practical experience with organizational alignment; 4) exploration of careers in QI/PI; and 5) time to plan and implement their own QI projects. Educational Activities: The course curriculum consisted of 4 weeks addressing advanced topics in QI/PI. Residents participated in over 80 Stanford Quality Improvement/Patient Safety Symposium 2015 Monday, June 8, 2015 at 3:30 PM – 6:30 PM Li Ka Shing Center (LKSC), Berg Hall, LK 230A Revised 06/04/2015 Page 17 of 21 hours of didactic and interactive activities including a 3-day QI/PI leadership seminar, lectures on core concepts, journal clubs, and a 3-day rapid process improvement workshop. In addition to faculty and peer mentorship on projects, active participation in hospital-prioritized improvement projects was a core feature of the curriculum. Assessment and Evaluation: Four residents participated. Confidence in QI/PI methodology was self-assessed pre and post course on a 5-point Likert scale (1=not at all confident, 5=extremely confident), and revealed improvements in the following: 1) responding to medical errors (2.25 vs 4.25), 2) writing a clear problem statement (1.75 vs 4.25), 3) determining appropriate QI/PI methodology (1.75 vs 3.75), 4) evaluating the effectiveness of a given methodological approach (2.00 versus 4.00), and 5) using the PDSA cycle (2.23 vs 4.25). All residents “strongly agreed” that they were “able to develop and implement a continuous quality improvement project” post-course (vs 25% pre-course), and 100% rated the overall course quality “outstanding.” Open response questions revealed hands-on QI/PI participation was the most beneficial to residents’ learning. 43. Timeout Compliance During Regional Anesthesia Training Jason Johns, MD; Luke McCage, MD*; Jean-Louis Horn, MD Department of Anesthesiology, Perioperative and Pain Medicine, Stanford University School of Medicine Introduction: The American Society of Regional Anesthesia and Pain Medicine (ASRA) recently published a pre-block safety checklist designed to be incorporated into the “timeout”. We evaluated the compliance to the standardized pre-block timeout by the individual trainees placing the block. Methods: One hundred surveys were filled out by the attending anesthesiologist supervising a resident or fellow performing a peripheral nerve block in the preoperative area. The primary question was whether or not the trainee performing the block initiated the mandated pre-block timeout. The trainees experience level was noted and the duration of the timeout pause was recorded. We calculated the percentage representing timeout compliance, analyzed compliance according to training level, and tabulated both the distribution and the duration of the timeout itself. Results: A total of 100 surveys were performed. On average trainees initiated a timeout 56% of the time. The amount of time spent to perform the timeout (80/100 surveys) was 71.2% taking less than 1 minute and 28.8% taking between 1-5 minutes. Resident timeout initiation during the rotation improved from an average of 41.5% during the first 2 weeks, to 66% the last 2 weeks while regional fellows averaged 75% throughout the 4 week period. Discussion: There is clear value in utilizing a safety checklist prior to block placement. In this case, the timeout was not initiated by the trainee 44% of the time, despite the quick nature of it. Overall, more emphasis needs to be placed on training providers to adhere to this valuable safety protocol. 44. Hydroxychloroquine (Plaquenil) and Retinal Toxicity; Yearly Eye Exam for Patients on Plaquenil Hooman Dehghan, MD, Neera Narang, MD Department of Rheumatology Purpose:  To ensure that all patients seen in the Division of Rheumatology and are on Plaquenil receive eye examinations at least yearly.  Also to ensure that all patients receive the recommended dose of Plaquenil (less than 6.5 mg/kg/day). Methods:  The Rheumatology division has reached a consensus regarding the importance of receiving an annual eye exam for patients who are using Plaquenil.  Reminded faculty/fellows regarding the importance of receiving an annual eye exam by routine discussion during clinics, reminder posters, pocket cards and re‐evaluation of Plaquenil dose for each patient.  Data was collected by reviewing electronic medical records.  Data was recorded and trended in the Rheumatology QI Dashboard.  Monthly confidential individual feedback was provided to each clinician via emails, EPIC messages and personal meetings regarding their performance as compared to that of the entire division. Results: Ongoing project. Results will be available at the end of May. Conclusions: Ongoing project. Results will be available at the end of May. 45. Non-mydriatic fundus camera screening for referral-warranted diabetic retinopathy in a Northern California safety-net setting Brian Toy*, James Egbert, Tyler Aguinaldo, Joseph Eliason, Darius Moshfeghi Department of Ophthalmology Purpose: Tele-ophthalmology fundus photography has shown promise in diabetic retinopathy screening. This study assessed the prevalence of diabetic retinopathy (DR) in a safety-net population in Santa Clara County, CA, based on non-mydriatic fundus photography screening, compared this method of screening with clinical dilated fundus examination, and evaluated patient and health systems factors associated with more severe DR. Methods: A retrospective chart review was conducted on 6,911 adult patients with diabetes, who presented to Santa Clara Valley Medical Center (SCVMC) for non-mydriatic fundus photography screening for DR between 2008 and 2012. Patient photos were graded for the presence of referral- warranted DR. A subset of 709 patients was referred to the SCVMC Eye Clinic for dilated fundus exam, and the clinical grade of diabetic retinopathy was compared with the non-mydriatic photo grade of DR. Results: Based on non-mydriatic screening, the prevalence of any DR was 17%, with moderate nonproliferative diabetic retinopathy (NPDR) or worse present in 5% of patients. 13% of photos were unreadable. When compared with clinical grading, the sensitivity and specificity of non-mydriatic grading to detect moderate NPDR or worse were 78% and 65%, respectively. Conclusions: Our study found that 17% of patients with diabetes in a large safety-net population screened with non-mydriatic fundus photography needed referral for further ophthalmic care. These results demonstrate a use-case of telemedicine to screen a large number of patients in a safety- net setting, but they also highlight the need for adequate specialty resources to care for patients referred for ophthalmic care. 46. A Quality Improvement Project to Reduce Continuous Pulse Oximetry Use in Pediatric Inpatients with Bronchiolitis Jody Lin, MD*; Yvette Keers-Moraga, RN; Jennifer Everhart, MD, Nivedita Srinivas, MD Department of Pediatrics Purpose: The 2014 AAP bronchiolitis guidelines recommend against routine continuous pulse oximetry (CPO) use in children admitted with bronchiolitis. Unnecessary CPO use is associated with alarm fatigue, prolonged hospitalization, and increased healthcare costs. We sought to Stanford Quality Improvement/Patient Safety Symposium 2015 Monday, June 8, 2015 at 3:30 PM – 6:30 PM Li Ka Shing Center (LKSC), Berg Hall, LK 230A Revised 06/04/2015 Page 20 of 21 Analysis/Results: Perioperative nurses record NPO violations in Epic and the Perioperative Analytics Team reports the incidence of violations. We assess the rate of NPO violations before and after the launch of the website to determine if these interventions decrease the rate of NPO violations. We are tracking the number of public visits to the NPO guideline page and will report correlation between webpage visits and incidence of NPO violations, assessing for significance. Conclusions: Our method involves a coordinated team approach between the Perioperative Analytics Team, Marketing, anesthesiologists, and proceduralists. This method of cross-specialty collaboration to decrease NPO violation will become increasingly important as the public’s desire for electronic information increases. Future interventions include the development of an app to send push email and text alerts to patients to further prepare patients for surgery. 54. Behavioral Analysis of Electronic Medical Record Use by Internal Medicine Residents on the General Medicine Service David Ouyang*, Jonathan Chen, Jason Hom, Jeffery Chi Department of Medicine Purpose: To understand the workflow and time constraints of internal medicine residents working on the general inpatient medicine wards at Stanford University Hospital. Methods: Retrospective analysis of resident usage of electronic medical record system use at Stanford University Hospital between June 26, 2013 and May 30, 2014. Results: We identified 4,327,708 unique actions performed by 87 residents during by their days of service on the general medicine wards. During this time, residents saw a median of 13 patients and worked 12.78 hours per day. Compared to supervising residents, interns looked at the chart of fewer patients (11.9 vs 25.8 patients, p < 0.001) and worked fewer hours ( 11.5 vs. 12.8 hrs, p < 0.001). We identify temporal patterns in computer use which correlate with known patterns in resident schedule including morning report, grand rounds, and noontime conferences. Housestaff frequently switch between tasks involving multiple patients suggesting a high degree of multiple tasking and possibly frequent interruptions. Conclusions: While researchers have sought to characterize resident work hours and workflows in this era of modern medicine, we introduce the analysis of electronic medical record system ultilization as the most granular and objective way to characterize resident behavior. In the future, we plan to use the data to assess whether team census affects resident workhours as well as patient outcomes. 55. Improving Care Coordination and Communication with Primary Care Providers in Single Ventricle Patients in the Interstage period Holly Bauser-Heaton, MD*; Bambi Alexander-Bays, PNP; Renee M. Rodriguez, MD; Bronwyn Harris, MD; Charles Nguyen, MD; Gail Wright, MD Divison of Cardiology, Department of Pediatrics Purpose: Single ventricle patients in the interstage period (between their initial surgical palliation and second stage) are amongst the highest risk patients. Our home monitoring program (HMP) team, consisting of physicians and nurse practitioners, closely monitors these patients but effective care coordination with primary care providers (PCPs) remains important . In collaboration with our Single Ventricle Family Advisory Council (FAC), we aimed to improve communication between the HMP team and PCPs. Methods: Based on results from a telephone survey of PCPs of recently discharged single ventricle patients, we developed an improved educational packet to be mailed prior to hospital discharge. The packet addressed identified gaps in communication. We continued to perform our pre-discharge phone call with PCPs. PCPs who received the new education packet then underwent the same telephone survey. Results: Post-surveys showed improvement in all areas identified during the pre-evaluation phase, including handoff accuracy (83% baseline, 100% post), receipt of the packet (50% baseline, 100% post), receipt of information about cardiac physiology (0% baseline, 83% post), and receipt of information about single ventricle care (0% baseline, 83% post). When PCPs were asked if the information was useful, the answer “Very/Somwehat” increased from 66% to 100%. Conclusion: Effective care coordination is especially challenging, yet imperative, in caring for complex patients. Continued evaluation and adjustment to communication practices, particularly at the time of hospital discharge, can improve PCP knowledge and provider relationships which is likely to result in improved patient outcomes. 56. Intraoperative medication safety - Exploring strategies to minimize medication waste and improve safety Shara Cohn, Amanda Kumar, Jed Cohn, Brita Mittal, Christina Stachur, Louise Wen, Clair Secomb, Ruth Fanning Department of Anesthesiology, Perioperative and Pain Medicine; Stanford School of Medicine Purpose: Anesthesiologists rely upon rapid access to essential drugs for emergent cases or unexpected intraoperative complications. Currently, anesthesia providers draw these drugs from vials into syringes to facilitate rapid administration if needed; however many are unused and discarded at the end of the day. Other workarounds include administering medications that have exceeded their shelf-life, storing medications in unauthorized locations, or passing medications to later shifts, promulgating expired medication use and possibly multiplying effects of a single medication error. Provider-based dilutions can increase risk of medication error, resulting in both patient safety and drug waste issues with potential economic consequences. The Anesthesia Patient Safety Foundation recommends that high alert drugs should be available in standardized concentrations in a ready-to-use form. We are exploring the option of compounded, sealed, pre-filled syringes as a safety and cost-saving measure. Methods: We performed a resident survey to evaluate which medications are routinely drawn up for anesthetic care, assess of amount of waste, and gauge resident concern regarding medication waste. Results: We surveyed Stanford CA-2 residents, and 100% of responders were concerned about drug waste in the OR. 90% reported >3 drawn but unused clean syringes wasted at the end of the day. 100% noted passing medications to other providers at the end of the day. Conclusions: Residents have a high level of concern for excessive drug waste. Further analysis is pending a planned trial in collaboration with Stanford OR pharmacy of pre-filled syringes of selected emergency drugs to compare cost savings and drug waste. 57. Nighttime Communication at Stanford University Hospital: Perceptions and Reality Andrew Sun MD, Minjoung Go MD, Gloria Sue MD, Erin Palm MD, Graeme Rosenberg MD, Raymond Deng, MSPH, Lisa Shieh MD*, Paul Maggio MD* Purpose: To improve physician, nursing, and care team communication. Methods: 236 pages to general surgery night float residents between 10/19/2014 and 10/25/14 and 11/2/14 and 11/5/14 were reviewed (9 night shifts total). Pages were categories by sender, urgency, and subject matter. Urgent pages were those pages that required an assessment or Stanford Quality Improvement/Patient Safety Symposium 2015 Monday, June 8, 2015 at 3:30 PM – 6:30 PM Li Ka Shing Center (LKSC), Berg Hall, LK 230A Revised 06/04/2015 Page 21 of 21 intervention by the night float resident. Non-urgent pages were those pages that did not require timely intervention and could be better addressed by the primary service. A paper survey for the nursing staff, and an electronic survey for the medicine and surgery residents were conducted to evaluate provider perception of nighttime communication. Results: On average, 26.2 pages were received each night by the night float resident. Of the 236 pages reviewed, 89% were sent by nursing staff. 12% were specified as “FYIs,” 39% requested an order, and 6% requested an MD evaluation. Categorization of the pages by urgency revealed 59% were urgent and 34% were non-urgent. 7% of pages could not be categorized. Of the urgent pages, 56% required a medication/procedure/order from the night float and 13% were related to abnormal vital signs. Of the non-urgent pages, 35% were non-urgent patient status updates and 21% were for low- priority medications/orders/procedures. Of 187 residents, 54 responded to the survey. 15% reported that they are satisfied or very satisfied with the current paging system. 22% reported that they are placed on hold for greater than 3 minutes. Resident perception, on average, is that 52% of nighttime pages are non-urgent or better addressed by the primary team. 154 of 200 distributed nursing surveys were completed. 54% stated that they are satisfied or very satisfied with the current paging system. 74 % stated that they wait at least 10 minutes to receive a response and that on average, 24% of pages are non- urgent. Conclusions: While the majority of pages sent to general surgery night floats are urgent (59%), a very significant proportion (34%) did not require an immediate response. This was supported by our survey and underscores the current lack of a non-urgent modality for communication.